pith. sign in

arxiv: 2606.24958 · v1 · pith:3C4BJ5W2new · submitted 2026-06-23 · 💻 cs.LG · cs.RO· math.DS

Swarm-Inspired Generation of Collective Behaviors in Graph Dynamical Systems

Ji Chen , Song Chen , Chengzhang Gong , Li Fan , Chao Xu This is my paper

Pith reviewed 2026-06-26 00:45 UTC · model grok-4.3

classification 💻 cs.LG cs.ROmath.DS
keywords collective synchronizationgraph dynamical systemssigned attentionswarm-inspired learningheterophilous graphsmulti-agent coordinationemergent synchronizationrobot locomotion control
0
0 comments X

The pith

SIES learns a generalizable coupling operator that produces prescribed synchronization patterns across untrained graph scales, phase relations, and node dynamics without retraining.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents SIES as a graph-dynamical framework that learns local interaction laws to generate desired collective behaviors such as synchronization. Signed source-target-conditioned attention serves as an adaptive coupling inside an explicit evolution model, allowing the same trained operator to control phase patterns on new networks and dynamics. The approach also transfers the signed interaction principle to message passing, yielding top results on heterophilous node classification. A sympathetic reader would care because the method aims to replace hand-designed or retrained controllers with one learned rule set that works across scales and tasks. This positions the framework for synchronization control, adaptive robot coordination, and heterophilous graph learning.

Core claim

SIES learns a generalizable coupling operator that produces prescribed synchronization patterns for CDSs across untrained network scales, target phase relations, and intrinsic node dynamics without retraining. The learned operator reaches gait-related modes faster than three oscillator baselines and generalizes synchronization-driven locomotion to simulated multi-legged robots of different scales and a physical hexapod after leg disablement. For graph representation learning, SIES applies the same signed interaction principle to message passing and achieves the highest performance among the compared methods on heterophilous node-classification benchmarks.

What carries the argument

signed source-target-conditioned attention acting as an adaptive coupling term inside an explicit evolution model

If this is right

  • Produces prescribed synchronization patterns for CDSs across untrained network scales, target phase relations, and intrinsic node dynamics without retraining.
  • Reaches gait-related modes faster than three oscillator baselines.
  • Generalizes synchronization-driven locomotion to simulated multi-legged robots of different scales and a physical hexapod after leg disablement.
  • Applies the signed interaction principle to message passing and achieves the highest performance on heterophilous node-classification benchmarks.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same signed coupling might be tested on other collective tasks such as flocking or consensus formation in multi-agent systems.
  • Replacing standard attention in existing GNN architectures with this signed version could improve results on additional heterophilous datasets beyond the benchmarks shown.
  • The explicit dynamical engine might allow direct incorporation of physical constraints when transferring the learned operator to real hardware beyond the hexapod example.

Load-bearing premise

The signed source-target-conditioned attention mechanism, when used as an adaptive coupling term inside an explicit evolution model, can be trained to produce interaction laws that generalize to new graphs, target patterns, node dynamics, and tasks without retraining or fine-tuning.

What would settle it

Training SIES on small graphs with one set of node dynamics and target phases, then testing whether it produces the prescribed synchronization on a graph with twice as many nodes, different intrinsic dynamics, and a new phase target.

read the original abstract

Collective behavior arises when locally interacting units produce coordinated global organization, from synchronization in dynamical systems to task-relevant information flow on graphs. The central challenge is not only to explain how collective behavior emerges, but to design local interaction rules that can produce desired global organization and generalize across graphs, dynamics and tasks.To address this challenge, we introduce the Swarm-Inspired Emergent Synchronizer (SIES), a graph-dynamical framework that learns generalizable local-interaction laws for controllable collective organization. Each node is an agent-like dynamical unit with a state and task cue, and signed source-target-conditioned attention acts as an adaptive coupling term inside an explicit evolution model. Therefore, SIES combines an explicit dynamical engine with local agent intelligence, similar to biological swarms. For synchronization control, SIES learns a generalizable coupling operator that produces prescribed synchronization patterns for CDSs across untrained network scales, target phase relations, and intrinsic node dynamics without retraining. The learned operator also reaches gait-related modes faster than three oscillator baselines and generalizes synchronization-driven locomotion to simulated multi-legged robots of different scales and a physical hexapod after leg disablement. For graph representation learning, SIES applies the same signed interaction principle to message passing and achieves the highest performance among the compared methods on heterophilous node-classification benchmarks. Together, these results position SIES as a generalizable and learnable graph-dynamical interaction framework with promise for synchronization control, adaptive robot coordination, and heterophilous graph representation learning.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 1 minor

Summary. The paper introduces the Swarm-Inspired Emergent Synchronizer (SIES), a graph-dynamical framework in which each node is an agent-like dynamical unit and signed source-target-conditioned attention serves as an adaptive coupling term inside an explicit evolution model. The central claims are that the learned coupling operator produces prescribed synchronization patterns for collective dynamical systems (CDSs) across untrained network scales, target phase relations, and intrinsic node dynamics without retraining; reaches gait-related modes faster than three oscillator baselines; generalizes synchronization-driven locomotion to simulated multi-legged robots of varying scales and a physical hexapod after leg disablement; and, when the same signed-interaction principle is applied to message passing, yields the highest performance among compared methods on heterophilous node-classification benchmarks.

Significance. If the generalization and performance claims are substantiated by the experiments, the work would offer a concrete bridge between explicit dynamical systems and learned local interaction rules, providing a reusable operator for controllable collective organization that applies across synchronization tasks, adaptive robotics, and heterophilous graph representation learning. The explicit evolution model plus agent-like attention is a distinctive architectural choice that could be reusable beyond the reported domains.

minor comments (1)
  1. The abstract asserts generalization without retraining and performance gains, but the provided text supplies no experimental details, validation procedures, error analysis, or data descriptions; the full manuscript must include these sections with quantitative results, ablation studies, and statistical significance tests to support the central claims.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their accurate summary of the SIES framework and for recognizing its potential significance as a bridge between explicit dynamical systems and learned local interaction rules. The recommendation of 'uncertain' is noted, but no specific major comments or concerns were provided in the report. Accordingly, we have no point-by-point responses to offer and no revisions to incorporate at this stage.

Circularity Check

0 steps flagged

No significant circularity identified

full rationale

The paper presents SIES as a learned graph-dynamical framework whose central claims rest on empirical training of a signed attention-based coupling operator followed by reported generalization tests across scales, dynamics, and tasks. No equations, derivations, or self-citations are exhibited in the provided text that reduce any prediction or uniqueness result to fitted inputs by construction. The approach is explicitly data-driven rather than analytic, with performance evaluated on held-out configurations, making the derivation chain self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides insufficient technical detail to enumerate free parameters, axioms, or invented entities; the framework introduces a new model architecture but no explicit parameter counts or background assumptions are stated.

pith-pipeline@v0.9.1-grok · 5808 in / 1261 out tokens · 28671 ms · 2026-06-26T00:45:05.967513+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

73 extracted references · 5 canonical work pages

  1. [1]

    Bonabeau, E., Dorigo, M., Theraulaz, G.: Swarm Intelligence: from Natural to Artificial Systems vol. 1. Oxford University Press, New York (1999) 19

    1999

  2. [2]

    Physical review letters75 6, 1226–1229 (1995)

    Vicsek, T., Czir´ ok, A., Ben-Jacob, E., Cohen, I., Sochet, O.: Novel type of phase transition in a system of self-driven particles. Physical review letters75 6, 1226–1229 (1995)

    1995

  3. [3]

    Scientific American269 6, 102–9 (1993)

    Strogatz, S.H., Stewart, I.: Coupled oscillators and biological synchronization. Scientific American269 6, 102–9 (1993)

    1993

  4. [4]

    PLoS Computational Biology16(4), 1007697 (2020)

    Rahmani, P., Peruani, F., Romanczuk, P.: Flocking in complex environments—attention trade-offs in collective information processing. PLoS Computational Biology16(4), 1007697 (2020)

    2020

  5. [5]

    PLoS Computational Biology15(9), 1007354 (2019)

    Heras, F.J., Romero-Ferrero, F., Hinz, R.C., Polavieja, G.G.: Deep attention networks reveal the rules of collective motion in zebrafish. PLoS Computational Biology15(9), 1007354 (2019)

    2019

  6. [6]

    Nature Communications15(1), 4779 (2024)

    Xiao, Y., Lei, X., Zheng, Z., Xiang, Y., Liu, Y.-Y., Peng, X.: Perception of motion salience shapes the emergence of collective motions. Nature Communications15(1), 4779 (2024)

    2024

  7. [7]

    Nature211, 562–564 (1966)

    Buck, J.B., Buck, E.: Biology of synchronous flashing of fireflies. Nature211, 562–564 (1966)

    1966

  8. [8]

    Nature Reviews Neuroscience19, 255–268 (2018)

    Muller, L.E., Chavane, F., Reynolds, J.H., Sejnowski, T.J.: Cortical travelling waves: mechanisms and computa- tional principles. Nature Reviews Neuroscience19, 255–268 (2018)

    2018

  9. [9]

    Nature355, 349–351 (1992)

    Davidenko, J.M., Pertsov, A.V., Salomonsz, R., Baxter, W.T., Jalife, J.: Stationary and drifting spiral waves of excitation in isolated cardiac muscle. Nature355, 349–351 (1992)

    1992

  10. [10]

    The European Physical Journal B61, 485–491 (2007)

    Filatrella, G., Nielsen, A.H., Pedersen, N.F.: Analysis of a power grid using a kuramoto-like model. The European Physical Journal B61, 485–491 (2007)

    2007

  11. [11]

    Proceedings of the National Academy of Sciences110, 2005–2010 (2012)

    D¨ orfler, F., Chertkov, M., Bullo, F.: Synchronization in complex oscillator networks and smart grids. Proceedings of the National Academy of Sciences110, 2005–2010 (2012)

    2005

  12. [12]

    In: International Conference on Learning Representations (2018).https://openreview.net/forum?id=rJXMpikCZ

    Veliˇ ckovi’c, P., Cucurull, G., Casanova, A., Romero, A., Li‘o, P., Bengio, Y.: Graph attention networks. In: International Conference on Learning Representations (2018).https://openreview.net/forum?id=rJXMpikCZ

    2018

  13. [13]

    Platonov, O., Kuznedelev, D., Diskin, M., Babenko, A., Prokhorenkova, L.: A critical look at the evaluation of gnns under heterophily: are we really making progress? ArXivabs/2302.11640(2023)

    arXiv 2023

  14. [14]

    Proceedings of the 2010 American Control Conference, 930–937 (2009)

    D¨ orfler, F., Bullo, F.: Synchronization and transient stability in power networks and non-uniform kuramoto oscillators. Proceedings of the 2010 American Control Conference, 930–937 (2009)

    2010

  15. [15]

    Nature Communications13(2021)

    Sajadi, A., Kenyon, R.W., Hodge, B.-M.S.: Synchronization in electric power networks with inherent heterogeneity up to 100% inverter-based renewable generation. Nature Communications13(2021)

    2021

  16. [16]

    Physiological Reviews55 2, 247–304 (1975)

    Grillner, S.: Locomotion in vertebrates: central mechanisms and reflex interaction. Physiological Reviews55 2, 247–304 (1975)

    1975

  17. [17]

    Science346(6206), 196–203 (2014)

    Ijspeert, A.J.: Biorobotics: Using robots to emulate and investigate agile locomotion. Science346(6206), 196–203 (2014)

    2014

  18. [18]

    Neural Networks 21(4), 642–653 (2008)

    Ijspeert, A.J.: Central pattern generators for locomotion control in animals and robots: a review. Neural Networks 21(4), 642–653 (2008)

    2008

  19. [19]

    Science Robotics9(89), 9579 (2024)

    Radosavovic, I., Xiao, T., Zhang, B., Darrell, T., Malik, J., Sreenath, K.: Real-world humanoid locomotion with reinforcement learning. Science Robotics9(89), 9579 (2024)

    2024

  20. [20]

    IET Cyber-Systems and Robotics3(1), 53–67 (2021)

    Chen, S., Liu, Y., Chen, T., Lou, J.: Rhythm motion control in bio-inspired fishtail based on central pattern generator. IET Cyber-Systems and Robotics3(1), 53–67 (2021)

    2021

  21. [21]

    ArXiv abs/2406.08993(2024)

    Luo, Y., Shi, L., Wu, X.-M.: Classic gnns are strong baselines: Reassessing gnns for node classification. ArXiv abs/2406.08993(2024)

    arXiv 2024

  22. [22]

    Journal of Nonlinear science3(1), 349–392 (1993) 20

    Collins, J.J., Stewart, I.N.: Coupled nonlinear oscillators and the symmetries of animal gaits. Journal of Nonlinear science3(1), 349–392 (1993) 20

    1993

  23. [23]

    primary gaits

    Buono, P.-L., Golubitsky, M.: Models of central pattern generators for quadruped locomotion i. primary gaits. Journal of Mathematical Biology42, 291–326 (2001)

    2001

  24. [24]

    SIAM Journal on Applied Dynamical Systems2(4), 609–646 (2003)

    Stewart, I., Golubitsky, M., Pivato, M.: Symmetry groupoids and patterns of synchrony in coupled cell networks. SIAM Journal on Applied Dynamical Systems2(4), 609–646 (2003)

    2003

  25. [25]

    SIAM Journal on Applied Dynamical Systems4(1), 78–100 (2005) https://doi.org/10.1137/040604711

    Golubitsky, M., Stewart, I., T”or”ok, A.: Patterns of synchrony in coupled cell networks with multiple arrows. SIAM Journal on Applied Dynamical Systems4(1), 78–100 (2005) https://doi.org/10.1137/040604711

    work page doi:10.1137/040604711 2005

  26. [26]

    PhD thesis, Lancaster University and University of Florence (February 2019)

    Lucas, M.: Synchronisation and stability in nonautonomous oscillatory systems. PhD thesis, Lancaster University and University of Florence (February 2019). https://maximelucas.github.io/assets/pdf/2019phdlucas.pdf

    2019

  27. [27]

    SIAM Journal on Scientific Computing31(4), 2706–2732 (2009)

    Wilkins, A.K., Tidor, B., White, J., Barton, P.I.: Sensitivity analysis for oscillating dynamical systems. SIAM Journal on Scientific Computing31(4), 2706–2732 (2009)

    2009

  28. [28]

    Nonlinear Dynamics113(8), 7827–7851 (2025)

    Chen, J., Fan, L., Xu, C.: Free gait transition and stable motion generation using CPG-based locomotion control for hexapod robots. Nonlinear Dynamics113(8), 7827–7851 (2025)

    2025

  29. [29]

    ArXivabs/2106.10934(2021)

    Chamberlain, B.P., Rowbottom, J., Gorinova, M.I., Webb, S., Rossi, E., Bronstein, M.M.: Grand: Graph neural diffusion. ArXivabs/2106.10934(2021)

    arXiv 2021

  30. [30]

    In: International Conference on Machine Learning (2022).https://api.semanticscholar.org/CorpusID:246608169

    Rusch, T.K., Chamberlain, B.P., Rowbottom, J., Mishra, S., Bronstein, M.M.: Graph-coupled oscillator networks. In: International Conference on Machine Learning (2022).https://api.semanticscholar.org/CorpusID:246608169

    2022

  31. [31]

    ArXivabs/2311.03260(2023)

    Nguyen, T., Nguyen, T.M., Honda, H., Sano, T., Nguyen, V., Nakamura, S.: From coupled oscillators to graph neural networks: Reducing over-smoothing via a kuramoto model-based approach. ArXivabs/2311.03260(2023)

    arXiv 2023

  32. [32]

    Reviews of Modern Physics91(4), 045002 (2019) https://doi.org/10.1103/ RevModPhys.91.045002

    Carleo, G., Cirac, I., Cranmer, K., Daudet, L., Schuld, M., Tishby, N., Vogt-Maranto, L., Zdeborov´ a, L.: Machine learning and the physical sciences. Reviews of Modern Physics91(4), 045002 (2019) https://doi.org/10.1103/ RevModPhys.91.045002

    2019

  33. [33]

    Physics-informed machine learning,

    Karniadakis, G.E., Kevrekidis, I.G., Lu, L., Perdikaris, P., Wang, S., Yang, L.: Physics-informed machine learning. Nature Reviews Physics3, 422–440 (2021) https://doi.org/10.1038/s42254-021-00314-5

    work page doi:10.1038/s42254-021-00314-5 2021

  34. [34]

    Nature , year = 2023, month = aug, volume =

    Wang, H.,et al.: Scientific discovery in the age of artificial intelligence. Nature620, 47–60 (2023) https://doi.org/ 10.1038/s41586-023-06221-2

    work page doi:10.1038/s41586-023-06221-2 2023

  35. [35]

    Nature Communications15(1), 10263 (2024)

    Chen, S., Liu, J., Wang, P., Xu, C., Cai, S., Chu, J.: Accelerated optimization in deep learning with a proportional- integral-derivative controller. Nature Communications15(1), 10263 (2024)

    2024

  36. [36]

    In: Robotics: Science and Systems, pp

    Righetti, L., Ijspeert, A.J.: Design methodologies for central pattern generators: An application to crawling humanoids. In: Robotics: Science and Systems, pp. 191–198 (2006)

    2006

  37. [37]

    2008 IEEE International Conference on Robotics and Automation, 819–824 (2008)

    Righetti, L., Ijspeert, A.J.: Pattern generators with sensory feedback for the control of quadruped locomotion. 2008 IEEE International Conference on Robotics and Automation, 819–824 (2008). IEEE

    2008

  38. [38]

    Science315(5817), 1416–1420 (2007)

    Ijspeert, A.J., Crespi, A., Ryczko, D., Cabelguen, J.-M.: From swimming to walking with a salamander robot driven by a spinal cord model. Science315(5817), 1416–1420 (2007)

    2007

  39. [39]

    IEEE Transactions on Industrial Electronics63(9), 5488–5500 (2016) https://doi.org/10.1109/TIE.2016.2577546

    Yu, H., Gao, H., Ding, L., Li, M., Deng, Z., Liu, G.: Gait generation with smooth transition using CPG-based locomotion control for hexapod walking robot. IEEE Transactions on Industrial Electronics63(9), 5488–5500 (2016) https://doi.org/10.1109/TIE.2016.2577546

    work page doi:10.1109/tie.2016.2577546 2016

  40. [40]

    ArXivabs/1609.02907 (2016)

    Kipf, T., Welling, M.: Semi-supervised classification with graph convolutional networks. ArXivabs/1609.02907 (2016)

    Pith/arXiv arXiv 2016

  41. [41]

    ArXivabs/1706.02216 (2017)

    Hamilton, W.L., Ying, Z., Leskovec, J.: Inductive representation learning on large graphs. ArXivabs/1706.02216 (2017)

    Pith/arXiv arXiv 2017

  42. [42]

    Collective dynamics of ‘small -world’ networks,

    Watts, D.J., Strogatz, S.H.: Collective dynamics of ‘small-world’ networks. Nature393(6684), 440–442 (1998) https://doi.org/10.1038/30918 21

    work page doi:10.1038/30918 1998

  43. [43]

    In: 2012 4th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob), pp

    Hoffman, K.L., Wood, R.J.: Turning gaits and optimal undulatory gaits for a modular centipede-inspired millirobot. In: 2012 4th IEEE RAS & EMBS International Conference on Biomedical Robotics and Biomechatronics (BioRob), pp. 1052–1059 (2012). IEEE

    2012

  44. [44]

    Scientific reports6(1), 30199 (2016)

    Aoi, S., Tanaka, T., Fujiki, S., Funato, T., Senda, K., Tsuchiya, K.: Advantage of straight walk instability in turning maneuver of multilegged locomotion: a robotics approach. Scientific reports6(1), 30199 (2016)

    2016

  45. [45]

    Journal of Experimental Biology226(4), 244688 (2023)

    Diaz, K., Erickson, E., Chong, B., Soto, D., Goldman, D.I.: Active and passive mechanics for rugose terrain traversal in centipedes. Journal of Experimental Biology226(4), 244688 (2023)

    2023

  46. [46]

    Nature521(7553), 503–507 (2015)

    Cully, A., Clune, J., Tarapore, D., Mouret, J.-B.: Robots that can adapt like animals. Nature521(7553), 503–507 (2015)

    2015

  47. [47]

    Science314(5802), 1118–1121 (2006)

    Bongard, J., Zykov, V., Lipson, H.: Resilient machines through continuous self-modeling. Science314(5802), 1118–1121 (2006)

    2006

  48. [48]

    Robotics6(4), 22 (2017)

    Zhang, T., Zhang, W., Gupta, M.M.: Resilient robots: Concept, review, and future directions. Robotics6(4), 22 (2017)

    2017

  49. [49]

    Scientific Reports7(1), 277 (2017)

    Owaki, D., Ishiguro, A.: A quadruped robot exhibiting spontaneous gait transitions from walking to trotting to galloping. Scientific Reports7(1), 277 (2017)

    2017

  50. [50]

    Science Robotics6(57), 6354 (2021)

    Thandiackal, R., Melo, K., Paez, L., Herault, J., Kano, T., Akiyama, K., Boyer, F., Ryczko, D., Ishiguro, A., Ijspeert, A.J.: Emergence of robust self-organized undulatory swimming based on local hydrodynamic force sensing. Science Robotics6(57), 6354 (2021)

    2021

  51. [51]

    Journal of the Atmospheric Sciences20(2), 130–141 (1963)

    Lorenz, E.N.: Deterministic nonperiodic flow. Journal of the Atmospheric Sciences20(2), 130–141 (1963)

    1963

  52. [52]

    Nature525(7567), 47–55 (2015)

    Bauer, P., Thorpe, A., Brunet, G.: The quiet revolution of numerical weather prediction. Nature525(7567), 47–55 (2015)

    2015

  53. [53]

    arXiv preprint arXiv:2202.07575 (2022)

    Keisler, R.: Forecasting global weather with graph neural networks. arXiv preprint arXiv:2202.07575 (2022)

    arXiv 2022

  54. [54]

    Science382(6677), 1416–1421 (2023)

    Lam, R., Sanchez-Gonzalez, A., Willson, M., Wirnsberger, P., Fortunato, M., Alet, F., Ravuri, S., Ewalds, T., Eaton-Rosen, Z., Hu, W., Merose, A., Hoyer, S., Holland, G., Vinyals, O., Stott, J., Pritzel, A., Mohamed, S., Battaglia, P.: Learning skillful medium-range global weather forecasting. Science382(6677), 1416–1421 (2023)

    2023

  55. [55]

    Wiley, ??? (2016)

    Balanis, C.A.: Antenna Theory: Analysis and Design, 4th edn. Wiley, ??? (2016)

    2016

  56. [56]

    Wiley, ??? (2009)

    Hansen, R.C.: Phased Array Antennas, 2nd edn. Wiley, ??? (2009)

    2009

  57. [57]

    Journal of Sound and Vibration392, 232–239 (2017)

    Borra, C., Pyles, C.S., Wetherton, B.A., Quinn, D.D., Rhoads, J.F.: The dynamics of large-scale arrays of coupled resonators. Journal of Sound and Vibration392, 232–239 (2017)

    2017

  58. [58]

    Nonlinear Dynamics111(20), 18585–18603 (2023)

    Borra, C., Bajaj, N., Rhoads, J.F., Quinn, D.D.: Multiple equilibrium states in large arrays of globally coupled resonators. Nonlinear Dynamics111(20), 18585–18603 (2023)

    2023

  59. [59]

    Hopf, E.: Abzweigung einer periodischen l¨ osung von einer station¨ aren l¨ osung eines differentialsystems. Ber. Math.- Phys. Kl S¨ achs. Akad. Wiss. Leipzig94, 1–22 (1942)

    1942

  60. [60]

    relaxation-oscillations

    Pol, B.: Lxxxviii. on “relaxation-oscillations” . The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science2(11), 978–992 (1926)

    1926

  61. [61]

    Progress in Mathematics, vol

    Golubitsky, M., Stewart, I.: The Symmetry Perspective: from Equilibrium to Chaos in Phase Space and Physical Space. Progress in Mathematics, vol. 200. Birkh¨ auser, Basel (2002)

    2002

  62. [62]

    Pinto, C.M., Rocha, D., Santos, C.P.: Hexapod robots: new CPG model for generation of trajectories. J. Numer. Anal. Ind. Appl. Math7(24), 15–26 (2012)

    2012

  63. [63]

    In: Springer Series in Synergetics (1984)

    Kuramoto, Y.: Chemical oscillations, waves, and turbulence. In: Springer Series in Synergetics (1984). https://api.semanticscholar.org/CorpusID:115807634 22

    1984

  64. [64]

    Hairer, E., Wanner, G.: Solving ordinary differential equations ii: Stiff and differential-algebraic problems. (2010). https://api.semanticscholar.org/CorpusID:117014104

    2010

  65. [65]

    ArXivabs/1909.13021(2019)

    Rozemberczki, B., Allen, C., Sarkar, R.: Multi-scale attributed node embedding. ArXivabs/1909.13021(2019)

    arXiv 1909

  66. [66]

    Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (2019)

    Akiba, T., Sano, S., Yanase, T., Ohta, T., Koyama, M.: Optuna: A next-generation hyperparameter optimization framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (2019)

    2019

  67. [67]

    arXiv preprint arXiv:1903.02428 (2019)

    Fey, M., Lenssen, J.E.: Fast graph representation learning with pytorch geometric. arXiv preprint arXiv:1903.02428 (2019)

    Pith/arXiv arXiv 1903

  68. [68]

    https://pytorch-geometric.readthedocs.io/en/latest/advanced/batching.html

    Advanced mini-batching technique. https://pytorch-geometric.readthedocs.io/en/latest/advanced/batching.html. Accessed: 2025-10-04

    2025

  69. [69]

    Proceedings of the 3rd International Conference on Vision, Image and Signal Processing (2019)

    Dankwa, S., Zheng, W.: Twin-delayed ddpg: A deep reinforcement learning technique to model a continuous movement of an intelligent robot agent. Proceedings of the 3rd International Conference on Vision, Image and Signal Processing (2019)

    2019

  70. [70]

    Pattern Recognit

    Fawcett, T.: An introduction to roc analysis. Pattern Recognit. Lett.27, 861–874 (2006)

    2006

  71. [71]

    Translated by AR Forrest (1972)

    B´ ezier, P.: Numerical control-mathematics and applications. Translated by AR Forrest (1972)

    1972

  72. [72]

    Pioneering and Leading Goose + X

    Phillips, G.M.: Bernstein Polynomials. Interpolation and Approximation by Polynomials. Springer, New York, NY (2003) Acknowledgments This work was supported by the “Pioneering and Leading Goose + X” Science and Technology Program of Zhejiang (Grant No. 2025C01052); the Key Research and Development Project of China National Tobacco Corporation (Grant No. 1...

    2003

  73. [73]

    For longer models, we choose larger stiffness coefficients to improve stability

    For shorter centipede models, we select smaller stiffness coefficients within this range to enhance turning flexibility. For longer models, we choose larger stiffness coefficients to improve stability. Supplementary Table 3: Key parameters of robots Physical robot Simulated robot Robot mass 14 kg Body Stiffness 1×10 −4∼5×10−4 Body mass 5.6 kg Segment mass...